Mutations resulting in disruption of the Gi cell cycle arrest pathways by loss of the INK4a-ARF gene, or mutations in p53 associated with cdk4 overexpression or Rb loss, are frequently found in glioblastomas. In addition, these tumors achieve activation of signal transduction pathways induced by tyrosine kinase receptors such as EGFR and PDGFR. These receptors activate a number of common pathways including those involving Ras, Src and AKT. Although it is known that these pathways are activated in gliomas, the role of their activity in gliomagenesis has not been fully understood. Our preliminary data shows that somatic cell gene transfer of activated forms of AKT and Ras, in combination, can induce glioblastoma formation in mice. Furthermore, expression of polyoma virus middle T antigen, known to activate the Ras and AKT pathways, also induces gliomagenesis. In these studies, we have used the RCAS/tv-a astrocyte-specific gene transfer system that involves avian retroviral vectors (RCAS) and mice expressing the RCAS receptor (tv-a) as a transgene from either the astrocyte-specific GFAP promoter or the glial progenitor-specific nestin promoter. This system allows combinations of genes carried on RCAS vectors to be transferred to astrocytes both in cell culture and by direct gene transfer in vivo. In this grant I propose to determine: (1) whether INK4a-ARF loss enhances glioblastoma formation in this assay, (2) if transformation is more efficient in glial progenitors than terminally differentiated astrocytes, and (3) what characteristics Ras and AKT induce in cultured mouse astrocytes.